Richard Ngo Accuses Effective Altruists of Corrupting AI Alignment Field
Reactions from ranked influencers
2 posts
You are applying an indefensibly low bar for “sincerely” here, in a way that leads to predictable mistakes about how bad outcomes arise—both in theory, c.f. our discussions about shadow; and in practice, c.f. the ways EAs corrupted the field of alignment, as recounted in this…
Thinking about this because I’ll soon put out part 2 of my retrospective on where alignment went wrong. It’d be easy to view it as vitriolic, but I hope I’ve managed to convey in it that I’m trying to evaluate others how I’d evaluate my past self in their shoes. Related:
@s_r_constantin @HiFromMichaelV @ben_r_hoffman @TheZvi Okay, well, every time I'm surprised by reality I look back and think "What about my model and my way of thinking could I change that would have predicted that better, without predicting a bunch of other things worse?"